Sangho Kim and Donghyun Park
Seminar at Research Institute of Economy, Trade and Industry (RIETI), Tokyo, 16 November 2011
Ownership structure and export performance:firm-level evidence from Korea
Introduction
A firm’s ownership structure affects a firm’s performance
Seminal paper by Berle and Means (1932)
Manager pursues own self-interest rather than owner interest in the absence of monitoring
With widely dispersed ownership, there is no incentive for monitoring
This implies that firms with more concentrated ownership will be more closely monitored
On the other hand, there are also theoretical bases for a negative relationship --- i.e. predatory owners
Introduction
Whether ownership concentration has a positive or negative impact on performance is ultimately an empirical question.
This is because there are theoretical grounds for both a positive and negative relationship
A large and growing empirical literature examines the issue.
Demsetz and Lehn (1985), Himmelberg, Hubbard and Palia (1999)
Overall evidence is mixed and inconclusive.
Introduction
The central objective of our paper is to empirically examine the relationship between ownership
structure and export performance.
Export performance is an important determinant of overall corporate performance.
But there are very few empirical studies which investigate the ownership structure-exports link.
Surprising and unfortunate
The main contribution of our paper is to help fill this gap in the literature using firm-level data from Korea.
Introduction
• Korea’s export success is driven by firms which vary widely in size and other structural characteristics
• Therefore, it is interesting to examine whether
ownership concentration has a significant effect on Korean firms’ export performance.
• Our study informs us about Korean firms’ exports.
• In addition, the study marks a first step toward filling a major gap in the literature.
Conceptual basis
• Entering the foreign market is a high-risk activity
• Sunk costs
• Revenue volatility due to exchange rate fluctuations
• Limited knowledge of local market conditions
• Often tougher competition
• Manager’s attitude toward risk influences a firm’s export decision-making.
• Ownership structure influences manager’s risk preference.
• Therefore, a firm’s ownership structure influences its export decision-making.
Conceptual basis
• The key link is the relationship ownership structure and risk preference.
• In a firm with dispersed ownership, manager maximizes his own expected utility rather than shareholders’
expected profits.
• In contrast, in a firm with concentrated ownership,
manager represents dominant shareholder’s interest and thus maximizes firm’s expected profits.
• Therefore, manager bears the risk of entering foreign markets.
• Bottom line: firms with concentrated ownership are more likely to export than firms with dispersed
ownership
Literature review
• There is a large empirical literature which looks at ownership structure-overall performance but…..
• ….. only a very limited literature on ownership structure and export performance.
• Most of the existing studies examine the link between foreign ownership and export performance.
• Cole et al (2010) use Thai manufacturing survey data.
• Other studies include Ngoc and Ramsetter (2009), Filatotchev et al (2008), and Wignaraja (2008).
• None of the studies look at other dimensions of ownership structure.
• In particular, they do not look at the impact of ownership concentration.
Data and empirical framework
• The data set consists of:
• Unbalanced panel of annual time-series for 463 Korean manufacturing firms during 1994-2005
• All manufacturing firms whose stocks are listed on Korea Stock Exchange
• Listed firms are required to report their financial status, and all our firm-level data are from the financial reports.
• Out of a total of 5,557 observations, exports are observed for 1,640 observations, or 29.5% of total
• We first build a binary variable of exporter/non- exporter
• We use the widely used logit model to regress the binary variable on a number of explanatory variables.
Data and empirical framework
• In the logit model, coefficient estimates indicate impact of explanatory variables on probability of being an
exporter.
• In addition, we use tobit model to study firm’s export propensity, which is ratio of exports to total sales.
• Our key variable is the ownership concentration rate
(CR), which is the ratio of dominant shareholder’s shares to total shares.
• We also control for other firm-specific factors widely used in the literature.
• These include wage rate, capital intensity, R&D stock, firm size, productivity, and firm age.
Data and empirical framework
• Wage rate = total labor costs/number of employees (L)
• Capital stock (K) = real amount of tangible fixed assets
• Capital intensity = capital/labor ratio = K/L
• Labor productivity = VA/L = per worker value added
• R&D stock = estimated by perpetual inventory methods based on R&D investment, depreciation rate = 10%
• Firm age is calculated from the founding year.
• All variables are converted into constant 2000 prices.
Empirical results: logit
• Table 3 reports the results.
• Most significantly, our estimation results show that higher ownership concentration rate increases the probability of a firm’s exporting.
• This suggests that firms with concentrated ownership
venture into risky export markets to maximize expected profits.
• On the other hand, firms with dispersed ownership tend to avoid risk and have a shorter term horizon.
• Our results imply the presence of an agency problem.
Empirical results: logit
• Table 3 results also indicate that firms are more likely to be exporters as wage rate, capital intensity, R&D stock, productivity and sales increase.
• Korean manufacturing exporters are larger, more capital- intensive, more R&D-intensive, and pay higher wages
than Korean manufacturing non-exporters.
• Our evidence is consistent with existing studies.
• Bernard and Jensen (1997), Aitken et al (1997), Clerides et al (1997), and others
• More efficient and larger firms with more resources are better able to cope with large sunk costs required for exporting.
Empirical results: tobit
• Export propensity, or ratio of export revenues to total sales, is the dependent variable.
• Export propensity is defined on [0, 1]
• Table 4 shows the coefficient estimates of the tobit estimation.
• To avoid biased estimates from applying OLS, we use a tobit model censored at both ends.
• As was the case for logit, our key variable of interest is ownership concentration ratio, and we also include the standard control variables.
• In estimation, we use a semi-log model to transform all explanatory variables into logarithms to control for
heteroscedasticity. Dependent variable is in original form
Empirical results: tobit
• Most significantly, Table 4 results indicate that an increase in the ownership concentration ratio boosts a firm’s
export performance.
• Firms with concentrated ownership are willing to bear the high level of risk required to enter a large number of markets.
• Exporting to distant, unfamiliar markets may boost profits and thus benefit shareholders.
• In contrast, firms with dispersed ownership may export only to fast-growing, geographically close markets.
• For example, Korean firms find China to be an attractive regardless of market structure.
• Our results imply the presence of an agency problem.
Empirical results: tobit
• The results for control variables are more or similar to the logit results.
• Firms with higher capital intensity, R&D stock and productivity have higher export propensity.
• One departure from logit results is that firms which pay lower wages have higher export propensity than firms which pay higher wages.
Concluding observations
• The central objective of our study was to empirically examine the relationship between the ownership
structure of firms and their export performance.
• To do so, we use firm-level data from Korean manufacturing sector
• It is intuitively plausible that ownership structure of firms has an effect on their export performance.
• Managers of firms with concentrated ownership have higher risk preference.
• We empirically test the ownership structure-export performance nexus using logit and tobit models.
• We find that firms with concentrated ownership export more than firms with dispersed ownership.
Concluding observations
• The main policy implication from our results is that policymakers need to take into account export
performance in policies on ownership concentration.
• In Korea, our results suggest the need for caution in policies which seek to regulate and control chaebols.
• There are a number of promising areas for future research.
• Examine ownership structure-exports nexus in other
countries, especially in East Asia and advanced economies.
• Examine ownership structure-outward FDI nexus.